The increasing use of clinical decision support systems (CDSS) to assist clinicians in decision-making is pushing the limits of information technology. The emergence of Electronic Health Records (EHR) coupled with enriched health information standards such as HL7 CDA, SNOMED, ICD-10 and LOINC have provided a rich environment for massive data collection and analysis by healthcare providers. This immense increase in data collection has also provided a gateway for the application of various data mining techniques on clinical datasets so as to measure health status (i.e. function, comfort and likelihood of dying) of patients. In measuring health status, many clinicians have opted to use CDSS to assist in decision-making and enhance clinical experience. However, even as the use of CDSS in clinicians' office continues to grow, the question that remains in the minds of many patients and the general public is whether it is appropriate, or ethical, for researchers to use health data collected for the purpose of direct patient care to develop computerized predictive decision support tool. In this paper, a systematic review is used to highlight the relevant technical barriers and ethical issues surrounding the secondary use of health data in developing CDSS.